Unauthenticated Python execution via unsandboxed model output
Published Aug 26, 2026 · Updated Aug 27, 2026
Eval injection in Google Langfun before 0.1.2 allows remote attackers to execute arbitrary Python code through crafted prompt input. The default lf.query Python protocol passes model-generated expressions to pyglove.coding.evaluate with sandbox disabled, while its AST filter permits calls without restricting callable names or builtins. An attacker needs influence over content that reaches the model prompt; successful injection runs with the host application's privileges and can read secrets, alter files, and issue network requests.
Summary
What happened
Eval injection in Google Langfun before 0.1.2 allows remote attackers to execute arbitrary Python code through crafted prompt input. The default lf.query Python protocol passes model-generated expressions to pyglove.coding.evaluate with sandbox disabled, while its AST filter permits calls without restricting callable names or builtins. An attacker needs influence over content that reaches the model prompt; successful injection runs with the host application's privileges and can read secrets, alter files, and issue network requests.
The record
- CVE
- CVE-2026-75062
- Published
- Aug 26, 2026
- Updated
- Aug 27, 2026
- Vendor
- Google Cloud
- Product
- Langfun
- Classifications
- CWE-95, CWE-1188, CAPEC-242, T1059.006
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Aug 26, 2026CVE publishedPublication date reported by the CVE source.
- Aug 27, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=0.0.1 <0.1.2
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- Langfun lf.query filesystem-canary reproductionproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo